258 A Leiler from M. Malus to the [April, 



position. I place in the direction of a polarized ray a pile of 

 parallel glass plates, and forming with its direction an angle of 

 35° 25', and 1 dispose them in such a manner relati-»'e to the poles 

 of the rays of ligiit that no light is reflected. I then make the 

 incident ray turn upon itself without changing its place, and pre- 

 serving the same inclination with respect to the pile. When it has 

 revolved the fourth part of the circumference, it is totally reflected 

 by the successive action of the plates, and it ceases to be perceived 

 at the extremity of the pile. After half a revolution it begins to 

 be transmitted again. Tlius when we perceive the greatest quantity 

 of refracted light, there is no reflected light ; and when we per- 

 ceive the greatest quantity of reflected light, there is no light re- 

 fracted beyond the pile. It is therefore very evident that the light 

 reflected in one case is transmitted in the other, and vice versa. 



It is not necessary to observe, that to turn a polarized ray on 

 itself I employ a ray formed by the ordinary refraction of a piece of 

 . Iceland crystal, the faces of which are parallel to each other, and 

 perpendicular to the direction of the ray. By turning round this 

 substance in its own plane, I change the direction of the poles of 

 the rays without varying their direction or intensity. 



I shall now state an observation which throws a new light upon 

 this subject. When an ordinary ray falls upon a plate of glass at 

 an angle of 35° 25', all the light reflected is polarized ; but the 

 transmitted ray contains (proportional to the rays reflected) a certain 

 quantity of light polarized in a direction diametrically opposite. If 

 we expose this ray to the successive action of different plates, the 

 quantity of transmitted light polarized accumulates incessantly : so 

 that, after a certain number of transmissions, all the light reflected 

 is polarized in one direction, and all the light refracted is polarized 

 in another. If we receive each of these rays upon rhomboids of 

 Iceland spar, the principal section of which is parallel to the plane 

 of incidence, the light reflected by the glasses is refracted in the 

 ordinary way, and the light transmitted, in the extraordinary way. 

 This has led me to the following conclusion, that as often as a 

 quantity of light is polarized in one direction an equal quantity is 

 polarised in the other. 



Metallic surfaces appeared to me to present the phenomena of 

 polarization in a very incomplete manner. In fact, the ray reflected 

 from them at all incidences is always susceptible of being divided 

 into two pencils by the reflection of Iceland crystal. We observe, 

 it is true, when the incidence is very great, that one of the images 

 becomes faint while the other increases in intensity; but the pheno- 

 menon is never sufliciently apparent to enable us to determine at 

 what angle it is a maximum. I have overcome that difliculty by the 

 following experiment. 



I receive upon a metallic mirror a ray already polarized, but so 

 that the plane of reflection, which passes through the incident and 

 reflected ray, makes an angle of 45° with the rectangular poles of 

 tlie ray. if the incidence is very small, or very great, the ray is 



